House Echo — 2026-07-21

desk: house-echo date: 2026-07-21 forecasts: echo: p_trade: 0.40 direction: none conviction: low surge: p_trade: 0.50 direction: long conviction: med

Desk Plan — House Echo — 2026-07-21

Briefing reference: /Users/dadbot/Desktop/ClaudeBod/projects/dadbrain/Analysis/briefings/2026-07-21-pm.md PM: ECHO (House Echo, Family 2 — Momentum)

Shared Market Read

• Event risk today: LOW — no macro catalysts. Clean session. The dominant narrative is pre-earnings positioning ahead of tomorrow's mega-cap earnings (Alphabet, Tesla after close) and next week's FOMC (July 29).
• Session character expected: Gap-up bias / modestly trending-up. Nasdaq futures leading on a chip-stock bounce (INTC +5.94%, TER +5.91%, STX +5.65% pre-market). Oil eased to $83.07 (+0.72%) on mediation efforts in the Middle East. The open-high-fade pattern from yesterday (SPY -0.65% from open, QQQ -0.91% from open) may or may not repeat — depends on whether the chip bounce has follow-through. Without a macro catalyst, the session is likely to track the broader pre-earnings positioning theme.
• VIX regime: 17.62 (LOW — <18). Dropped back into LOW regime after yesterday's MID close (18.65). For SURGE: VIX < 20 cohort PF 1.36 — the weakest VIX band for SURGE's edge. For ECHO: LOW VIX is neutral for the vol_ratio gate; the gate tripped on 7/17 at VIX 18.08 (SPY vol_ratio 1.2392) but today's 17.62 is a lower-vol baseline.
• Key levels:
• SPY: Prior close 742.15. SMA20 744.74. Prior range 741.56–748.69. Below SMA20 for 2nd consecutive day. The 740 area is near-term support.
• QQQ: Prior close 695.97. SMA20 716.04. Prior range 695.53–705.78. 4th consecutive day below SMA20. 20pts below — widest divergence in this cycle. The 700 level is the most watched technical level.
• SOXX: 535 make-or-break level — untested but important. The chip rout pause yesterday (Nvidia +2.39%) is the stabilizing signal.
• XLE: Oil at $83.07. Energy rotation active on Hormuz tensions. EIA petroleum report is tomorrow (July 22) — not today.
• Macro backdrop: The chip rout (PHLX Semi index -18% from June peak) is the dominant unresolved narrative, but the selloff is pausing. The correction is orderly, not panicked (VIX 17.62). The two-tier divergence (tech sold off, rotation into energy/cyclicals) continues. Pre-earnings positioning is the day's operational theme.

ECHO — Close-Leg Exhaustion Trader (Dynamic Universe: SPY/XLE/XBI)

Setup present: possible — more likely than yesterday. The chip bounce creates a clear morning direction candidate (up), which is the precondition ECHO needs. The gap-up bias from the Asian rally and chip-stock bounce gives the first hour a plausible directional impulse. The key question is volume: will the pre-earnings positioning generate enough institutional flow to push vol_ratio above 1.2?

Vol_ratio gate — the binding constraint. After tripping on 7/17 (SPY 1.2392 on VIX 18.08 gap-down) and being untested on 7/20 (no gap, no catalyst), today's profile is different: a catalyst-driven gap-up (chip bounce) with Asian-market volume behind it. The first hour's institutional volume on a gap-up day with a clear sector narrative (semiconductors) is structurally higher than yesterday's catalyst-less Monday. The vol_ratio gate is more likely to trip today than on 7/20, but the question is whether the 20-session average (which includes the high-volume 7/17 session) shifts the ratio unfavorably.

Morning direction assessment: The overnight chip bounce (INTC +5.94%, TER +5.91%) creates a strong bullish bias for QQQ and the broad market. SPY should follow with a moderate gap-up. The key question is whether the gap-up holds or fades. If it holds:

• Morning direction: UP (bullish)
• ECHO would enter SHORT at 15:30 (fading the exhausted morning rally)
• The exhaustion thesis: chip bounce is a relief rally, not a trend reversal — institutional buying may exhaust by 15:20

If the gap-up fades (like yesterday's open-high-fade pattern where SPY opened near 748 and closed at 742):

• Morning direction: could be DOWN (if the fade is sharp enough that the first-hour return is negative)
• ECHO would enter LONG at 15:30
• The exhaustion thesis: the chip bounce failed, sellers reasserted, and by 15:20 the selling is exhausted

XLE watch: The strongest ECHO instrument (61.3% WR, 3.22 PF in backtest). Oil at $83.07, still elevated on Hormuz tensions. The energy rotation — the desk's most active sector story — continues. EIA is tomorrow, not today, so there's no scheduled catalyst. But the ongoing rotation provides a structural directional bias. If XLE gaps up with oil, it's a strong SHORT candidate for ECHO at 15:30 (fading the energy rotation).

Exhaustion_score concern: Even if vol_ratio trips, the exhaustion_score (vol_ratio × |signal_return|) needs to clear 0.80. On a gap-up day, the signal return could be 0.50%–0.80% — at vol_ratio 1.2, that's 0.60–0.96. The score is borderline. On a fade day with a smaller signal magnitude (0.30%–0.50%), the score will struggle even if vol_ratio trips. The exhaustion_score floor is the quiet binding gate today.

Call: Possible, but the vol_ratio and exhaustion_score gates are the binding constraints. The chip bounce gives the session a clearer directional character than yesterday, which is the precondition. The most plausible candidate is SHORT (fading the chip bounce relief rally) or XLE SHORT (fading the energy rotation). I rate this at p_trade 0.40 — more likely than yesterday's 0.20, but still below even odds.

Key invalidation: If the chip bounce fades immediately (gap-up opens and reverses within the first 30 minutes), the morning direction is ambiguous — ECHO's worst case. A clean fade that establishes a clear DOWN direction by 10:00 is actually better for the exhaustion thesis (LONG at 15:30) because it creates a cleaner signal. The worst case is a two-sided, choppy morning with no clear direction.

SURGE — Mid-Session Momentum Continuation (VWAP Pullback, SPY/QQQ)

Setup present: more likely than yesterday — the best setup SURGE has seen in the last several sessions. Three factors align today that were absent or hostile yesterday:

1. Clear directional impulse from the chip bounce. QQQ has a genuine catalyst-driven gap-up bias (Nasdaq futures leading, INTC +5.94%, TER +5.91%, STX +5.65%). This is a cleaner first-hour signal than yesterday's 0.23% futures lead. The first-hour |signal_return| for QQQ should clear the ±0.30% gate with room to spare. SPY is a secondary candidate — the broad-market bounce is less pronounced than the sector-specific chip bounce.
2. Clean entry window — no event in the window. Unlike yesterday's 10:00 AM LEI, there is no scheduled catalyst in SURGE's 10:00–12:00 ET entry window. The three-phase structure can form without disruption. The gap-up provides the starting impulse (Phase 1: trend establishment in the first hour). The natural pullback to VWAP (Phase 2) is the classic pattern for a gap-up day that doesn't immediately run away.
3. VWAP pullback is the natural pattern for a gap-up day. The classic gap-up scenario: open above VWAP, first-hour consolidation/drift, measured pullback to VWAP (often between 10:00–11:00 ET), then resumption (Phase 3). This is the exact structure SURGE's strategy was designed to capture. The pre-earnings positioning theme (tomorrow's GOOGL/TSLA earnings) provides a reason for the rally to have follow-through — institutions positioning ahead of mega-cap earnings.

VIX concern — the dampener. VIX at 17.62 (LOW regime) is SURGE's weakest VIX band. The backtest shows VIX < 20 cohort PF 1.36 vs. VIX 20–30 PF 2.18. The edge is structurally weaker in LOW VIX. This is the primary reason I rate this as "med" conviction rather than "high" — the setup is structurally present, but the edge is diminished in the current VIX regime.

ADX gate: The chip bounce should produce a clean directional first hour, which will build ADX. On a trending session with a gap-up, ADX ≥ 22 is very likely. The TEMPER finding that moderate-trend (ADX 22–30) outperforms strong-trend (ADX ≥ 30) is relevant — if the session is modestly trending rather than explosively so, that's actually favorable for SURGE's edge.

Direction assessment: LONG. The chip bounce is the dominant narrative. QQQ has the strongest catalyst (semiconductor names leading the pre-market). SPY follows. The pre-earnings positioning theme (Alphabet, Tesla tomorrow) gives the rally a reason to persist. The risk is that the gap-up fades quickly (like yesterday's open-high-fade pattern) — if the first hour is negative, the direction flips, and the chip bounce narrative is disproven. But the base case is a positive first hour.

Instrument selection (co-fire limiter): QQQ has the stronger structural thesis — the chip bounce, 20pts below SMA20, the widest divergence in this cycle. A bounce from oversold conditions on a sector-level catalyst (semiconductors) is the cleanest setup. SPY is a secondary candidate — the broad-market bounce is less pronounced. If both qualify, favor QQQ LONG.

Call: Likely — the best setup SURGE has seen since the 7/17 gap-down session. The chip bounce provides a clear directional impulse, the entry window is clean, and the VWAP pullback structure is the natural anatomy of a gap-up day. The VIX regime (LOW, 17.62) is the dampening factor — the strategy's edge is weaker here. I rate this at p_trade 0.50 with LONG direction.

Key invalidation: If the gap-up fades immediately (open near 700+ and drops below 695 within the first hour), the bullish thesis is invalidated. In that case, if the first hour is decisively bearish (|signal| ≥ -0.30%), SURGE could enter SHORT — betting the chip bounce was a relief rally that failed. But the base case is LONG. The other invalidation: if the gap-up runs too far and never pulls back to VWAP (gap-and-go), the VWAP pullback structure cannot form. This is less likely — gaps typically produce a pullback in the first 90 minutes.

Plan Filed

• Filed: 2026-07-21 pre-market
• Frontmatter forecasts complete for every active member: yes
• Primary desk call: SURGE is the more likely trade today — the gap-up, clean entry window, and VWAP pullback pattern align for the first time in several sessions. ECHO is possible but constrained by the vol_ratio gate and exhaustion_score. The chip bounce is the day's defining narrative — it gives SURGE a setup to trade and ECHO a direction to fade.
• What I'm watching for a mid-session reassessment: (1) Whether the gap-up holds or fades in the first hour — this determines SURGE's direction and ECHO's morning signal. (2) Volume profile in the first hour — if vol_ratio is trending toward 1.2, ECHO's assessment improves. (3) SOXX 535 level — a clean bounce from this level validates the chip bounce narrative; a break below invalidates it.
• TEMPER awareness: Today is a clean test of SURGE's VWAP pullback setup in a LOW VIX regime with a gap-up catalyst. This is a regime condition the backtest shows weaker PF (1.36) — a trade today would be a live data point in that calibration. For ECHO, today is another vol_ratio test in a no-gap catalyst environment (the chip bounce is a catalyst, but for QQQ, not SPY/XLE/XBI directly).

House Echo Desk Reflection — 2026-07-21

Desk: house-echo (Family 2 — Momentum / Trend Continuation) PM: ECHO Period covered: 2026-07-11 (desk creation) through 2026-07-21 — the desk's first full reflection cycle, triggered by SURGE's winning trade today. Members: ECHO (contra-trend, 15:30 ET), SURGE (with-trend, 10:00–12:00 ET)

Desk-Level Summary

Combined desk P&L: 7 trades, 6 W / 1 L, +$974.44

The desk enters its first reflection period with both stacks in positive territory and both SPRT monitors in CONTINUE with comfortable boundary distances. Neither stack has accumulated enough trades for statistical evaluation — 3 and 4 trades respectively are below any Phase 3 checkpoint — but the directional signal is constructive. Both members have fired at least once in the LOW VIX regime that the backtest identified as their structurally weakest environment, which is itself a live data point worth monitoring rather than a problem requiring action.

Sessions by Day (desk creation to today)

ECHO — Close-Leg Exhaustion (Contra-Trend, 15:30 ET)

What Happened (Full Period)

3 trades taken since Phase 3 began, 3 wins, 100% WR, +$110.71 cumulative. All three were short-duration time_exit wins — XLE short (7/07, +$77.61), XLE short (7/13, +$26.40), SPY long (7/17, +$6.70). The mean win of ~$37 reflects the tight intraday range of the close window.

The 100% WR is a statistical artifact of small n (3). The backtest projects 56.4% dev WR and 73.1% holdout WR — 3-for-3 is within sampling noise and should not anchor expectations.

Binding Gate Analysis: vol_ratio ≥ 1.2 — LEANING REGIME

ECHO's gate stack requires ADX ≥ 20, vol_ratio ≥ 1.2, VWAP persistence, AND exhaustion_score ≥ 0.80. Across ~11 non-event, non-skip sessions since Phase 3 began, ECHO has fired on 3 (27%). Of the 8 no-trade sessions, the binding gate has been vol_ratio on the majority.

The pattern: ECHO fires when a clear directional program with institutional volume is present (7/07: XLE gap-down -0.33% with energy rotation volume; 7/13: oil/energy pullback volume; 7/17: gap-down with above-average volume at VIX 18.08). On narrower-range, lower-volume days — which characterize the current LOW VIX regime (VIX 17–18) — vol_ratio struggles to clear 1.2.

This is regime, not calibration. The vol_ratio ≥ 1.2 threshold was calibrated in the 2021–2023 dev period which included the 2022 bear market (high vol) and 2023 recovery (moderate vol). The current July 2026 environment — VIX 17, normal-range trending days at 0.95–1.00× ATR, no macro catalysts through 7/28 — is structurally less voluminous. The near-miss data (where available, the gap between measured vol_ratio and 1.2 is material, not a hair) suggests the setup is genuinely absent, not that the threshold is marginally wrong.

What I'm watching: If the VIX regime shifts to MID (20–30) in August — next Tier-1 trigger is FOMC 7/29 — the vol_ratio gate becomes easier to clear and we'll learn whether ECHO's firing rate recovers to dev-projected levels. If VIX stays LOW through August, the live firing rate deviates meaningfully from backtest and that's a TEMPER-relevant data point, not a unilateral tweak.

Plan Calibration

On the two days where direction was forecast (short on 7/15, 7/16, long on 7/17), the single trade day (7/17) correctly anticipated direction. The no-trade forecasts for 7/20 (p_trade 0.20) were accurate. The 7/21 forecast of p_trade 0.40 was optimistic — the vol_ratio gate didn't clear. 3/5 Brier scores are good (0.04–0.16), 2/5 are elevated (0.09–0.25). Too small a sample to draw calibration conclusions.

SURGE — Mid-Session Momentum Continuation (VWAP Pullback, SPY/QQQ)

What Happened (Full Period)

4 trades taken since Phase 3 began: 3 wins, 1 loss, 75% WR, +$863.73 cumulative. The track record: SPY short +$68.85 (7/08, VWAP_BREAK), QQQ long -$43.77 (7/09, VWAP_BREAK), QQQ short +$326.60 (7/13, HARD_FLAT), SPY long +$512.05 (7/21, HARD_FLAT). Both winning intraday momentum trades (7/13, 7/21) were clean continuation sessions that ran all day. Both VWAP_BREAK exits (7/08 win, 7/09 loss) reflect the strategy's sensitivity to sessions where the VWAP structure breaks before the hard flat at 14:00.

The 7/21 SPY long was the highest single P&L — a textbook execution of the three-phase structure: chip bounce gap-up (Phase 1), retracement to VWAP at 745.72 at 10:10 (Phase 2), resumption holding through 14:00 (Phase 3). Exit at 748.42 produced a clean $2.70/share × 38 shares win.

Binding Gate Analysis: Catalyst Presence — LEANING REGIME

In the LOW VIX regime (VIX < 20, PF 1.36 per backtest), SURGE's edge is structurally weakest. The strategy requires a clean three-phase structure: trend establishment in the first hour, a controlled pullback to VWAP with contracting volume, then resumption. In the LOW VIX environment, the absence of a sector-level or macro catalyst produces directionless sessions where the first hour establishes no clear trend, and the three-phase structure cannot form.

The pattern across no-trade days (7/10, 7/15, 7/16, 7/17, 7/20) is consistent: no catalyst → no directional impulse → no VWAP pullback structure. The days that produced trades (7/08, 7/09, 7/13, 7/21) all had a catalyst: 7/08 (SPY gap-down continuation), 7/09 (QQQ gap-up tech bounce), 7/13 (QQQ gave down on broad market weakness), 7/21 (chip bounce — Advantest +8.13%, Nikkei +3.36%).

This is regime, not calibration. The catalyst-driven gap-up days are a distinct subset of the LOW VIX session universe. On days without a catalyst, the setup genuinely does not approach qualification. The near-miss data is not about a gate threshold being marginally too tight — the precondition (clean three-phase trend structure) is absent. A calibration tweak (e.g., relaxing the pullback depth or the resumption confirmation) would import lower-quality setups, which is the precise failure mode the plan's "med conviction, not high" recognized.

Event-Day Skip Diagnostic (completed 2026-07-17)

The WS2c Tier 1 diagnostic was resolved during this period: MIXED. Event-day trades are profitable in both periods (dev PF 1.54, holdout PF 12.11), but FOMC is a consistent loser while QQQ_EARN and PCE are consistently strong. Blanket skip softened but not changed — no live-path action taken.

ADX Re-Tune Watch (open)

TEMPER's supplemental finding that moderate-trend (ADX 22–30, PF 2.14) outperforms strong-trend (ADX ≥ 30, PF 1.65) remains untested in production. Today's trade (7/21 SPY) was trending but not explosive — may have fallen in the moderate-trend band. Worth logging ADX at entry for SPRT sub-cohort analysis.

2023 PF 0.95 Watch (open)

Rolling 6-month PF monitor built into paper_trade.py. Current 4-trade track record (75% WR, ~3.5× win/loss ratio implied) suggests healthy early trajectory but is too small to evaluate against the P12 trigger (PF < 1.2 for two consecutive windows). No action.

Plan Calibration

The 7/17 forecast is the notable miss: called short on a session that was ultimately +0.15% SPY. The gap-down from 7/16 (-0.26%) carried a bearish bias but didn't continue. The 7/21 forecast was the best — correctly identified the chip bounce catalyst, VWAP pullback structure, and LONG direction at p_trade 0.50. The direction hit rate (4/5 no-trade windows avoided false calls; 1/1 trade days correct direction) is strong at this small sample.

Open TEMPER Items (Carried from SURGE v9)

1. Event-day skip diagnostic: ✅ RESOLVED 2026-07-17 — MIXED result, no live-path change.
2. ADX re-tune question: ⏳ Open — moderate-trend (ADX 22–30) outperforms strong-trend (ADX ≥ 30) in backtest. Monitor live ADX at entry for sub-cohort analysis.
3. 2023 PF 0.95 watch: ⏳ Open — 4 trades is too small to evaluate the rolling 6-month PF monitor. Continue monitoring.

Cross-Stack Observations (Desk-Blind)

Complementary coverage is working as designed. The desk's structure — SURGE entering with-trend at 10:00, ECHO entering contra-trend at 15:30 — means they respond to different session conditions. On a clean trending-up day like 7/21, SURGE's setup fires while ECHO's does not (the trend was clean, not exhausted, so the exhaustion score at 15:20 was below threshold). On mixed/transition days like 7/17, ECHO's contra-trend setup fired while SURGE's with-trend setup did not (the direction was uncertain enough that no clean three-phase structure formed).

The 7/13 session was the only day both traded — QQQ gave down convincingly, establishing a clear DOWN trend. SURGE shorted QQQ at 11:50 (with-trend, +$326.60); ECHO shorted XLE at 15:30 (fading the exhausted oil rotation, +$26.40). Both won, and they addressed different instruments (QQQ vs. XLE) — no position overlap.

Both stacks operate at 0.75% dollar risk per trade, so the desk's total exposure on co-fire days is 1.5% — well within the combined risk envelope. Co-fire remains rare (1 of 7 trades across both stacks), which is appropriate behavior for complementary (not redundant) strategies.

Desk-Level Call

• Regime diagnosis: Both stacks are operating in their structurally weakest VIX regime (LOW, < 20). The reduced firing rate for both is consistent with the backtest's VIX regime breakdowns, not a calibration error. No action recommended.
• Next catalyst: FOMC on 7/29 is the next Tier-1 event. SURGE hard-skips FOMC; ECHO hard-skips FOMC. Both stacks will be inactive that session. The post-FOMC drift (7/30 onward) may shift VIX regime to MID, which would improve both stacks' structural environment.
• Firing rate monitoring: At current rates (ECHO: ~3 trades/month, SURGE: ~4 trades/month), the first Phase 3 checkpoint for SPRT confidence requires ~12 more weeks of data for ECHO and ~9 more weeks for SURGE. No statistical conclusions should be drawn until those checkpoints.
• Learnings flag: None this reflection — both stacks are performing within backtest-expected bounds, and the regime diagnosis (LOW VIX firing-rate suppression) is consistent with existing TEMPER findings (SURGE's VIX < 20 PF 1.36 cohort, ECHO's vol_ratio dependency). The data is accumulating as specified.

Next Reflection Trigger

Per WS5: same-day-on-trade for SURGE (the desk's higher-frequency stack). Next reflection is due on SURGE's 5th trade — expected in ~2–3 weeks at current cadence.

Filed: 2026-07-21 post-close